Hydra is a high-performance Postgres database with serverless online analytical processing (OLAP). It is designed for low-latency applications built on time series and event data.

• Homepage

The benchmark itself requires psql to execute. You will need to install psql, e.g.:

# Debian, Ubuntu, etc
apt-get install postgresql-client
# MacOS
brew install postgresql

Note that either option below will introduce additional latency depending on the location from which the benchmark is run.

In this configuration, you will run a Postgres instance locally and it will connect to Hydra's serverless OLAP service. This will use the Hydra CLI which will manage a Docker container for you.

1. Install Python, and if desired, your preferred Python management tools, like pipx, uv, etc.

2. Install Hydra CLI using pip install hydra-cli (or via pipx, uv, etc)

3. Install Docker.

4. Sign up for Hydra at https://start.hydra.so/get-started and copy the token to your clipboard

5. Run hydra setup and paste in your token

6. Run hydra start to start the local instance

7. Run hydra config to obtain the connection string and set it DATABASE_URL in your local environment, or pass it in when you run the benchmark.

8. Run the benchmark.

   DATABASE_URL="..." ./benchmark.sh
   

9. Run hydra stop to stop the local instance

In this configuration, you will connect to a Postgres instance already running in the cloud which is already configured to connect to Hydra's serverless OLAP service.

1. Sign up for Hydra at https://start.hydra.so

2. Request access and complete onboarding

3. Create a project

4. Obtain the connection string and set it as DATABASE_URL in your local environment, or pass it in when you run the benchmark.

5. Run the benchmark.

   DATABASE_URL="..." ./benchmark.sh